Unleashing Fusion Power: The Reality of ITER in the 21st Century

[peter0302]

Ok, all you people in the know, what's the real low-down on ITER and the possibility of fusion power reactors this century?

If you believe wikipedia, and ITER themselves, the theory behind efficient fusion power generation is sound and the problems that exist are engineering problems (such as, how do we contain the plasma, how do we prevent the neutrons from weakening the housing, etc.)

Is a fusion power reactor definitely going to happen, the only question being when? If so, why are the US and EU not spending a trillion dollars to do whatever it takes to accelerate the progress like we did for the Manhattan Project? Clearly fusion is the answer to all the world's energy problems - the fuel is cheap, there's no radioactive waste, and the power generated is enormous (10 times the heat put in, according to ITER).

Why isn't fusion the obvious answer?

 

[humanino]

If you believe wikipedia, and ITER themselves, the theory behind efficient fusion power generation is sound and the problems that exist are engineering problems (such as, how do we contain the plasma, how do we prevent the neutrons from weakening the housing, etc.)

Well, we made fusion weapons right So the theory seems to be right.

If so, why are the US and EU not spending a trillion dollars to do whatever it takes to accelerate the progress like we did for the Manhattan Project? Clearly fusion is the answer to all the world's energy problems - the fuel is cheap, there's no radioactive waste, and the power generated is enormous (10 times the heat put in, according to ITER).

Why isn't fusion the obvious answer?

I agree with you, and I think some people are doing their best to make it happen. Why we don't do more is a matter of political debate. Fusion would certainly not solve the oil issues for instance.

 

[peter0302]

Right right. What I wanted to make sure was that there were no serious theoretical problems with the idea of an efficient fusion power reactor (such as, for example, it takes more energy to start the fusion fire than you get out of it) or that there was a legitimate fear that it just plain ol' wouldn't work.

Also there are fundamental differences between fusion power and thermonuclear bombs (besides the obvious). (Forgive me if I'm stating what people already know). The big bombs usually use a three-stage (fission-fusion-fission) process. The 1st stage is like an ordinary A-bomb - very inefficient, i.e., most of the fissile material doesn't react. But it generates the heat necessary to make the deuterium/tritium plasma and start the fusion "fire". The fusion fire then starts a second fission reaction but this one many times more efficient than the first, and so most (if not nearly all) of the energy actually still comes from fission, and either way, they all rely on fission igniters. So it's not obvoius that just because the bombs work that clean fusion with no fission could be made efficient.

Leaving all that aside, though, if the only issues are political at this point, that's pretty pathetic. The U.S. spent $2 billion on the Manhattan Project in the 1940s ($20-something billion in 2008 dollars), but is spending less than $1 billion on something that would end the climate crisis, starvation, and water shortages in one swoop, not to mention revolutionize space travel. And yet we probably spend a billion per day on oil.

 

[taylaron]

Here here. I have yet to become a nuclear physicist, but I agree that even with my skimpy knowledge of fusion, It is possible; heck, that is all our sun does...

We (the people) need to put enough money and resources into this effort to figure out a feasible way to give it some momentum. money and resources. just like peter said, its like the Manhattan project. except one thing,

Its about saving the planet, not just the USA...

I do agree that it will not solve our fuel for transportation crisis, but it is an essential part of our near future.

It would certainly be a nice thing if this was something that our next President would keep in mind...
do you not agree?

 

[Astronuc]

Why isn't fusion the obvious answer?

Well, that's what ITER is supposed to demonstrate, at least for controlled thermonuclear reaction (CTR) using magnetic confinement, as opposed to inertial confinement. Then there is the Z-pinch system being researched.

Government funding for such programs has to compete with all the other programs government supports, e.g. the military actions in Iraq and Afghanistan (~$167 billion this year), Social Security, Medicare/Medicaid/, . . . .

In Fiscal Year 2006, the U. S. Government spent $406 Billion of taxpayers money on interest payments* to the holders of the National Debt. Compare that to NASA at $15 Billion, Education at $61 Billion, and Department of Transportation at $56 Billion.

from - http://www.federalbudget.com/ (see the Dept. of Energy expenditures)

But let's focus on the technical matters.


The fuel is not exactly inexpensive - deuterium is not abundant, and tritium must be made, either by neutron capture in deuterium and Li. The tritium and deuterium must be collected and supplied to the fusion plant. The d+t reaction is the easiest fusion reaction we can achieve. Then there is d+d, but that requires slighly higher temperatures (and pressures, unless plasma density is reduced).

The d+t reaction releases a 14.1 MeV neutron (or 80% of the 17.6 MeV released from the d+t fusion). The d+d reaction releases a 3.5 MeV neutron (and He-3 ion) in about 50% of reactions, and proton and triton in the other 50%. The He-3 is nice if it fuses with d (d + He3 => p + He-4), i.e. it's aneutronic so all the energy is released as charged particles.

There are still many technical challenges. For example, the neutrons from fusion will irradiated the containment structure and energy conversion system, which over time will become radioactive, and which will experience some degradation of mechanical properties, which is a safety as well as performance issue. So fusion does produce some radioactive waste.

Then there is energy conversion. Besides the tradiational thermal energy conversion to mechanical to electrical (usually with a 30-40% efficiency), are there more efficient extraction methods such as direct energy conversion. That remains to be seen.

We now await the proof of the ITER concept, and that's about 10 years away.

Detailed plans exist for the construction, operation and decommissioning of ITER, and indicate that, if the ITER Organisation is established in 2006, the first plasma should be possible in ITER by the end of 2016.

http://www.iter.org/a/n1/timeline.htm


The sun does fusion (p-p chain reaction with some CNO-cycle) under conditions that mankind cannot reproduce in a controlled manner. The energy densities and pressures are well beyond our technological capability.


Controlled thermonuclear fusion is not a slam, dunk!

 

[peter0302]

Thanks, Astronuc. That was precisely the kind of technical information I was looking for. I was most surprised by your answer that the fuel isn't cheap. For some reason I was under the impression that deuterium was abundant in ocean water. Is that wrong?

The biggest technical issue seems to me the irradiation of the containment chamber (and making it radioactive is also potentially very bad). Why would the neutrons not be magnetically contained just as the plasma? Too energetic?

Also how would the electricity actually be generated? Would the heat from the reaction power turbines?

Again I'm just trying to figure out if there are any potential show-stoppers or if it's all just a matter of some innovative engineering.

 

[Haelfix]

Fusion has always been more of an engineering problem than a physics one. There are some regimes of the plasma that are still a little bit out of analytic control, but by and large what we need are engineering breakthroughs and simply more research money to test things out.

Its most physicists opinion that we will see a working Fusion reactor, capable of semi sustained burns, within our lifetimes. As to the next step, actual energy efficient and cost effective manafacturing solutions, well that's yet another can of worms. The jury is out on that (and its way outside my scope of knowledge)

 

[Astronuc]

Thanks, Astronuc. That was precisely the kind of technical information I was looking for. I was most surprised by your answer that the fuel isn't cheap. For some reason I was under the impression that deuterium was abundant in ocean water. Is that wrong?

Well, seawater is abundant, and there is deuterium in it, but there is a cost associated with extracting the deuterium, which apparently has a concentration on the order of 16-20 ppm in seawater (depending on the source).
See this discussion - http://hyperphysics.phy-astr.gsu.edu/Hbase/nucene/fusion.html#c6
(I need verify the numbers because I've found conflicting numbers)

The biggest technical issue seems to me the irradiation of the containment chamber (and making it radioactive is also potentially very bad). Why would the neutrons not be magnetically contained just as the plasma? Too energetic?

Neutrons are of course neutral, so they do not strongly interact with magnetic fields, i.e. they cannot be contained by magnetic (or electric) fields. In fact, neutral atoms will drift/diffuse out of magnetic fields, which is another issue (hydrogen absorption in the first wall).

Irradiation and activation of the superconducting magnets will also be an issue. Neutrons disturb the atomic structure in materials, and decay of activated nuclides changes the chemical (elemental) properties of a material.

Also how would the electricity actually be generated? Would the heat from the reaction power turbines?

There are several ways. The conventional (or more traditional) way is to use the thermal energy from fusion (transferred through one or more working fluids), which is ultimately passed through a turbine, which drives an electrical generator. Ideally one would like to some how use the forces/pressures from the plasma, i.e. extracting work directly from an expanding plasma, or an MHD approach, using charge separation where the electrons go to one electrode and the nuclei to another electrode. The electrons provide a current as the travel through a conductor and then neutralize the nuclei being collected at the other electrode.

Again I'm just trying to figure out if there are any potential show-stoppers or if it's all just a matter of some innovative engineering.

As Haelfix indicated, the perfection of fusion is considered more of an engineering (applied physics) problem, but understanding the physics (behavior and characteristics) of plasmas is key. Then we are constrained by the technical limits (strength, creep) of the materials and the ability to produce strong and steady (controlled) magnetic fields. We just have to be very clever.

 

[peter0302]

Again thank you.

So one final question. Say I'm the President, and you're my science advisor, and I say, Astronuc, we need to save the earth, and I'm ready to gamble on fusion. 70 years ago Roosevelt gave the Army a blank check and the best minds in the country to make the atom bomb. If I give you the same, can you promise me you guys can make this thing work?

 

[Vanadium 50]

can you promise me you guys can make this thing work?

They couldn't even promise that they would make the atom bomb work. There are no guarantees in R&D - and even if you make it work, there's no guarantee that it will be accepted by the outside world. AT&T had "Picturephone" demoed in the 1964 World's Fair - but I still don't have one on my desk.

 

[humanino]

Say I'm the President, and you're my science advisor Astronuc

:uhh:

we need to save the earth, and I'm ready to gamble on fusion. 70 years ago Roosevelt gave the Army a blank check and the best minds in the country to make the atom bomb. If I give you the same, can you promise me you guys can make this thing work?

President, if you give me a blank check, I'll do it
:tongue2:

There are no guarantees in R&D

Yes, I think that needs to be realized by some politicians. When you have two projects, one of them saying "I need $100M and 3 years with my 5k top-researcher and engineer team" and the other says "We'll do it with $5M in 6 months with the 50 students and luck", politician be better think twice who they're going to give it to.

If it has never been done, it's research
If it has been done once, it's engineering
If it has been done twice, it's technical

AT&T had "Picturephone" demoed in the 1964 World's Fair - but I still don't have one on my desk.

You don't !? How much would pay for one ?

 

[Astronuc]

So one final question. Say I'm the President, and you're my science advisor, and I say, Astronuc,

Things are that desperate?

BTW, humanino, I appreciate the vote of confidence. :rofl: I'd have the same reaction.

we need to save the earth,

dire even!

years ago Roosevelt gave the Army a blank check and the best minds in the country to make the atom bomb. If I give you the same, can you promise me you guys can make this thing work?

I wouldn't ask for a blank check. At this point, I'd wait for the results of ITER to see what is achieved.

One concern I have about big projects is the expectation that they continue even when goals are not met. The other concern is - how big?

There are some other threads on fusion that would be worthwhile to review.

 

[humanino]

<off topic>

BTW, humanino, I appreciate the vote of confidence. :rofl: I'd have the same reaction.

I'm not so much concerned about the science advisor than I am about the president
A honest science advisor can always ask other scientists when he needs decision making.
Is it "An honest" ?
</off topic>

 

[Astronuc]

<off topic>I'm not so much concerned about the science advisor than I am about the president
A honest science advisor can always ask other scientists when he needs decision making.
Is it "An honest" ?
</off topic>

'An honest' <silent 'h'>.

Honesty and integrity are essential, even vital, for science. I remember back in the days of SDI or 'Star Wars' about lofty promises made by some notable scientists who were advising president Reagan. I think big egos and $$$$$ clouded sound judgement. That's the human side of science.

I would like to bring the discussion back to the technical issues, such as choice of fuel, confinement, power input (e.g. microwave, neutral beam injection, ohmic heating, . . .), and power extraction. I think the ITER site has information on each topic, and various threads scattered around PF address some of these aspects.

 

[peter0302]

I think I'd do a better job than the current one, at least.

Good point about the a-bomb being a gamble. But we took it. Is the fusion gamble a worse, equal, or better bet than the manhattan project? I think the need is just as dire, if not more so, and the potnetial return is definitley more attractive (this gamble might save the earth, whereas the other one nearly destroyed it).

But interesting that Astronuc would wait for ITER to start showing results. It just seems so far off - 2050 - to have a working reactor. I have to believe that if more money and resources were dedicated to this, it would happen a lot faster. Probably could happen before any new oil from off-shore rigs started hitting the gas pump!

 

[humanino]

<off topic>

'An honest' <silent 'h'>.

Thanks

I think I'd do a better job than the current one, at least.

Agreed, but that is not sufficient a criterium :rofl:
</off topic>

I would like to bring the discussion back to the technical issues, such as choice of fuel, confinement, power input (e.g. microwave, neutral beam injection, ohmic heating, . . .), and power extraction. I think the ITER site has information on each topic, and various threads scattered around PF address some of these aspects.

ITER site
They indicate 11.6 T maximum magnetic field. Is there anything special about this value being determined by confinement properties, or could efficiency increase if we were able to have higher magnetic field ? By that I mean, was 11.6 the result of a long evaluation of output versus total input ? I did not find much information on power extraction.

 

[Astronuc]

Good point about the a-bomb being a gamble. But we took it. Is the fusion gamble a worse, equal, or better bet than the manhattan project? I think the need is just as dire, if not more so, and the potnetial return is definitley more attractive (this gamble might save the earth, whereas the other one nearly destroyed it).

Controlled fusion is more difficult to achieve than fission. In fission, one could use high explosives (implosion) and the gun device to achieve critical mass - with relatively high purity of the fissile nuclides Pu-239 and U-235, respectively.

In fusion the reactions are pretty straighforward and well understood. The scientific and technical challenge in fusion is to achieve the conditions under which CTR may occur, and then to the perfect the process to extract the surplus energy, bearing in mind that some fusion energy is required to maintin the thermal conditions in the fusion reactor.

But interesting that Astronuc would wait for ITER to start showing results. It just seems so far off - 2050 - to have a working reactor.

In theory, ITER is the next stage in the evolution of CTR, based on previous experiments with the other systems like those at Princeton, General Atomics, Culham and others.

http://www.pppl.gov/fusion_basics/pages/fusion_basics.html
http://fusion.gat.com/global/Home
http://www.jet.efda.org/

 

[Orion1]

Astronuc, the isotopic neutron in artificial Tritium fuel only originates from a fission reaction, therefore in order to supply a full power fusion reactor with Tritium fuel, would still require a full power fission reactor to generate the Tritium fuel?

Artificial Tritium is produced in nuclear reactors by neutron activation of lithium-6, which has a natural abundance of only 7.5%, which means that it is a very limited natural resource.

Also, according to what I have studied about 'Tokamak' type fusion reactors are the problems with instabilities in thermonuclear plasma turbulence. Thermonuclear plasmas are very turbulent and very unstable and to my knowledge, these problems still have not been completely solved.

 

[parsec]

Neutrons are of course neutral, so they do not strongly interact with magnetic fields, i.e. they cannot be contained by magnetic (or electric) fields. In fact, neutral atoms will drift/diffuse out of magnetic fields, which is another issue (hydrogen absorption in the first wall).

Don't neutrons have a magnetic moment? Presumably high energy neutrons are still difficult to contain.

 

[Astronuc]

Astronuc, the isotopic neutron in artificial Tritium fuel only originates from a fission reaction, therefore in order to supply a full power fusion reactor with Tritium fuel, would still require a full power fission reactor to generate the Tritium fuel?

Artificial Tritium is produced in nuclear reactors by neutron activation of lithium-6, which has a natural abundance of only 7.5%, which means that it is a very limited natural resource.

More generally, one needs a 'practical' neutron source, which these days is a fission reactor. The other approach would be to use neutrons from the d+t or d+d reaction to produce tritium in a Li-6 blanket. That still doesn't address the fact that Li-6 is a finite resource. d+d also produces t ( d + d => t + p), but in theory, the t would be consumed in the fusion plasma.

Some tritium is also produced in CANDU reactors and LWRs through n-capture by d's, but the cross-section for that reaction is quite low.

Also, according to what I have studied about 'Tokamak' type fusion reactors are the problems with instabilities in thermonuclear plasma turbulence. Thermonuclear plasmas are very turbulent and very unstable and to my knowledge, these problems still have not been completely solved.

Yes, plasma instabilities are a problem and have been a challenge for magnetic confinement. Hopefully the magnetic confinement systems has been designed with these in mind. That will be part of the demonstration.

Don't neutrons have a magnetic moment? Presumably high energy neutrons are still difficult to contain.

Yes. The interaction of the neutron magnetic moment is very weak compared to the momentum of a 1 MeV neutron, or even eV neutrons.

Neutrons in typical plasma have a mean free path on the order of 10⁸ m, which means they'll leave the plasma and interact with the surrounding structure.

 

[peter0302]

Not that you haven't answered enough questions yet (and thank you for that ), but why isn't p+p possible in a controlled environment? Are the temperatures and pressures necessary simply too high?

 

[gdp]

Not that you haven't answered enough questions yet (and thank you for that ), but why isn't p+p possible in a controlled environment? Are the temperatures and pressures necessary simply too high?

The problem is that this reaction also requires the "Weak Force," not just the "Strong (Nuclear) Force," p + p => Deuteron + positron + neutrino --- and the "Weak Force" is so incredibly weak that the cross-section for this reaction (crudely speaking, the probability that the reaction will "go" per unit flux of incoming particles) is very small indeed --- for example, even under "solar interior" conditions, the rate at which this reaction "goes" is about 1 per proton-pair-collision per 14 BILLION years! (About the only reason we can "observe" this reaction in the Sun at all is that it is the "bottleneck" reaction that sets the net rate for all subsequent reactions, plus, the Sun contains a LOT or protons, and it has a LOT of time...)

In fact, the reaction cross-section for p+p fusion is so small that it has never been directly measured in the laboratory (or even once directly observed!); the best that lab experiments can currently do is provide extremely weak upper bounds. The p+p fusion cross-section has therefore so far only been either inferred or estimated, e.g., via theoretical calculations, or by indirect means such as working "backwards" from stellar theory, or by invoking "crossing symmetry" on related reactions that can be observed in the lab, e.g., Deuteron + neutrino => p + p + electron.

For technical details on the current state of solar physics, see http://www.sns.ias.edu/~jnb/Papers/Preprints/nuclearfusion.html.

 

[Astronuc]

Not that you haven't answered enough questions yet (and thank you for that ), but why isn't p+p possible in a controlled environment? Are the temperatures and pressures necessary simply too high?

Pressure p and temperature, T, go hand in hand, by virtue of p = nkT, where n is the particle density, k is Boltzmann's constant and T is the particle temperature.

The sun has conditions that are impossible to achieve in magnetic confinement on earth.

The core of the sun has conditions (http://cassini.mps.ohio-state.edu/~pogge/Ast162/Unit2/structure.html#HSE):

T = 15 Million K
Density = 150 g/cc (that's 150 times the density of water at room temperture - and it's at 15 million K).
P ~ 200 billion atm or 20.3 million GPa, or 20.3 billion MPa, or ~2.9 trillion psia


See also - http://csep10.phys.utk.edu/astr162/lect/energy/ppchain.html
http://hyperphysics.phy-astr.gsu.edu/hbase/astro/procyc.html

http://ircamera.as.arizona.edu/NatSci102/lectures/suninterior.htm

http://www.nasa.gov/worldbook/sun_worldbook.html

This puts the sun in perspective - http://hyperphysics.phy-astr.gsu.edu/hbase/solar/sun.html


Some nontechnical information on ITER: EU gets tough on fusion reactor
http://news.bbc.co.uk/2/hi/science/nature/4044895.stm